STHD: probabilistic cell typing of Single spots in whole Transcriptome spatial data with High Definition

STHD: probabilistic cell typing of Single spots in whole Transcriptome spatial data with High Definition

• Machine learning model for subcellular cell type prediction for high-resolution, high gene coverage spatial transcriptomics data, such as VisiumHD from 10X Genomics.
• Preprint: https://www.biorxiv.org/content/10.1101/2024.06.20.599803 
• GitHub: https://github.com/yi-zhang/STHD 
• Generates single-spot (2um) cell type labels and probabilities for VisiumHD data using a machine learning model.
• Input: VisiumHD data and reference scRNA-seq dataset with cell type annotation.
• Output: cell type labels and probabilities at 2um spot level.
• Visualization - STHDviewer: interactive, scalable, and fast spatial plot of spot cell type labels, in a HTML.

• Author: Yi Zhang, PhD, yi.zhang@duke.edu
• Website: Yi Zhang Lab at Duke
• STHDviewer of VisiumHD colon cancer sample with near 9 million spots: STHDviewer_colon_cancer_HD:https://yi-zhang-compbio-lab.github.io/STHDviewer_colon_cancer_hd